[ofw] [Fwd: Connection teardown in WinSock Direct and CMrelatedquestions]

[Sean Hefty]

>> The Windows stack is designed around having the CM perform the QP state
>> transitions.
>
>Do you know if the Windows CM implementation matches this design ?

I believe that this is the case, but I'm not familiar enough with the code to
know for sure.  It does appear that way based on a casual inspection of the
code.

>> The situation that you mentioned above (transitioning the QP into
>> ERROR when sending the DREQ) can result in a race condition where outstanding
>> sends *from* the remote side can timeout, even though the data's been
>received
>> by the local side (the one disconnecting).
>
>So sounds like that workaround in WinSock direct should be removed
>then ?

I would think that you'd want to wait for a DREQ sent timeout, DREQ received, or
DREP received before transitioning the QP into the error state.  Fab's on
vacation this week, but he may have better input next week when he returns.
Otherwise, maybe Tzachi knows, since I believe he's the WSD maintainer.

- Sean